Potentially accessible information is produced in the digital world at an ever increasing volume. Problems in the ability to deliver digital information securely to the right place, at the right time, and at an affordable price inhibit the utility and diminish the value of information for viewing and processing purposes. There is a need for businesses and their customers to access information in near real-time from any location on the globe in a secure, private, and cost effective manner.
The standards, technologies, and resulting network infrastructure presented by the Internet have established the value of sharing information and demonstrated the effect that a common infrastructure can have in driving low cost information sharing in a network environment. The Internet does this in a loose governance model where security, reliability, privacy, and enforcement are insufficient to assure the ability to confidently share mission critical or high value data. There is a need for an Internet like network implementation where the benefits of the Internet such as low cost, flexibility, reach, shared access and use, user control, multi-functionality, and the like can be accomplished. Current drawbacks of the Internet include, inter alia, the lack of quality of service, rouge user attacks (hackers), the absence of standards for global authentication and authorization, the lack of service level enforcement and tools for governance, the lack of standard application implementation and reporting, and the lack of assured delivery and status reporting. As a result, there is a need for an Internet like network solution that provides the benefits of the Internet such as flexibility, dynamics, and end user controllability, without the foregoing drawbacks.
Information sharing relationships can generally categorized as one-to-one, one-to-many, or many-to-many. Information sharing participants may have one or many of such relationships whether the participant is sharing information as an individual or a business. As such, each participant desires to maintain the confidentiality of relationships whether the relationships are driven by personal or business motives. In addition, all participants desire to exchange information in a secure and reliable manner. Traditional information and application sharing methods, such as dedicated high cost point-to-point network connections or industry-specific communities, have proven to be technically complex, have limited functionality, have limited reach, and are cost prohibitive for many applications. Point-to-point connections work well for a small set of isolated relationships, but as the number of relationships and interconnections increase, a point-to-point approach becomes too complex and inflexible to be practical. Industry specific communities typically provide a limited ability to reach outside the boundaries of a specific industry. The secure service implementation described herein as used on a common low cost network infrastructure could provide significant business value. Such a design will lower the participation cost for all parties while allowing a participant to consume services on an as needed basis. In addition, service providers can publish services for consumption by users as defined by the service provider independent of a dedicated network community or consortium.
Policy-based management is an existing approach used in an enterprise or network. Policies are operating rules established for situations that are likely to occur, such as for controlling access and establishing priorities for the use of resources. Unlike point-to-point management where security devices are configured one by one across the network to attain an appropriate security level, policy-based management closely follows business practices and requirements by establishing rules and relations among and between network entities. An entity defines policies for connections, access and applications accessible in the network. Policies are rules, independent of brand or function, propagated across devices, infrastructure or participant communities. Administrators do not need to know the specific language of any particular brand of equipment to set security rules according to business practices.
Current policy implementations must be set with firewalls in mind. Firewalls are hardware or software security implementations that provide a user with a protected network control by securing traffic at the network level, commonly referred to as the Internet Protocol (IP) stack. Firewalls are commonly used to provide a secure transition to the Internet, to separate an organization's public servers from the rest of the network, and to maintain individual network segments isolated and secure from one another. Firewalls can employ several techniques in combination to provide protection: basic packet filtering based on an IP address and/or port numbers (essentially a routing function), proxy or application-level servers that close the direct path between networks, network address translation (often used in proxy servers), content filtering, and DoS detection. These features are native to any modern day IP network implementation.
Firewalls traditionally enforce security at the data packet level within the network IP stack, namely, the firewall has little or no knowledge of the application or data content, or of the participants, except that participants are an anonymous user on an IP based network. As a result, all security related to authentication and authorization as well as reporting and logging are left to the end user or application developer for design and implementation. This results in an inconsistent implementation that does not easily and reliably support a trusted user community—let alone private user communities on a common network infrastructure. There is thus a need to establish a common service infrastructure that supports a higher level of commonality on a standard IP network. This higher level of commonality is focused on addressing the shortfalls in the current Internet model so that any to any secure and reliable connectivity can be accomplished in a cost effective and unrestrictive fashion.
A virtual private network (VPN) is a secure method of accessing a private network using a public network, such as the Internet. A remote user or network connects to a local Internet service provider (ISP), and then accesses a central site through a secure VPN mechanism. VPN's use data encryption to ensure that the data is secure from snooping. A VPN implementation can result in significant cost savings when compared to the expense of leased lines or dial-up connections for remote users. VPN security means are typically transient, existing only for the duration of the information exchange, and establish a virtual circuit across the Internet by encapsulating the original IP packets within specially secured IP packets.
A firewall can allow VPN traffic to pass back and forth; some firewalls are able to initiate and terminate VPN tunnels. This allows multiple PCs on a LAN to share the same VPN tunnel, and can speed up the performance of the VPN tunnel if the firewall uses hardware to encrypt the data. VPN tunneling protocols enable encapsulation of various protocols over an IP network and typically require authentication, authorization, and/or encryption. Content filtering can be included as part of a firewall or other security device. Content filtering limits a user's web site access by associating a restriction policy with a specific URL, IP address, or topic category. Content filtering can utilize a database containing a comprehensive list of elements to be blocked by limiting access to data associated with a specific IP address.
Authentication methods are used in most environments, such as face-to-face activities, over the phone, and/or Internet transactions. Current authentication technologies and techniques use a static design structure that is difficult to change and/or enhance. A typical authentication protocol involves a hub and spoke design and may violate privacy legislation dictates. Current authentication techniques involve a high participation cost because of the equipment, connectivity, and other requirements and provide only a single offering solution. Many authentication techniques require unique network requirements for each application or service offering.
To access a domain, a user must be authenticated and authorized to see that domain; a network participant must be a member of the domain or be in a trust relationship to the domain. To access resources on another domain, one domain sets up a trust relationship with the other domain. Two types of trust relationships exist: one-way trusts and bi-directional trusts. A trust relationship allows a participant to access another domain as though it is part of the participant's domain. Trust relationships often support coarse grained partitioning of rights and privileges. For example a domain may include all servers associated with a specific business unit such as human resources (HR). As a result of being an authorized member of the HR network domain, a participant/user can see all the HR computers that are part of that domain; however, it is likely that the applications on the servers will require user authentication and authorization at each application to access any data the applications may provide. This means that a user needs a user ID and password for each application in the domain. As the domain and the number of applications grow, the need to remember and manage a great number of user ID and password credentials may become overwhelming. Access control is a method of allowing and disallowing certain operations on a network or in an application. Access control may be implemented by access control lists (ACL). Access control typically supports finer grained domain partitioning allowing domain requesters to be authorized to perform a limited set of activities that may be available in a given domain.
Although many of these individual technologies exist, no solution is sufficiently adaptable to provide a secure service network offering that addresses the shortcomings, while maintaining the benefits of, the Internet. No known solution allows a service network capable of sharing sensitive information between and among participants in real time using a shared IP infrastructure. The benefit of a secured service network environment is significant in that the security available in a secure service network would support information sharing in a manner not currently available.